Status of the Ganges River Dolphin Or Shushuk Platanista Gangetica in Kaptai Lake and the Southern Rivers of Bangladesh

Oryx Vol 35 No 1 January 2001

Status of the Ganges river dolphin or shushuk Platanista gangetica in Kaptai Lake and the southern rivers of Bangladesh

Brian D. Smith, Benazir Ahmed, Muhammad Edrise Ali and Gill Braulik

Abstract Surveys for shushuks Platanista gangetica extend into the Bay of Bengal. No dolphins were were conducted during January to April 1999 in Kaptai observed in Kaptai Lake, a dam-created reservoir of Lake and the southern rivers of Bangladesh. A popula- the upper Karnaphuli, despite reports of occasional tion of at least 125 dolphins was recorded in the sightings by local fishermen. No shushuks were ob- Karnaphuli and Sangu rivers and connecting canal. served in the Bagkhali and Matamuhuri rivers, possibly The overall encounter rate- was 0.76 dolphins per km. because of seasonal-closure dams present near the Density was highest in the lower reaches of the Sangu, mouths of both rivers. The main threats to dolphin where we recorded 1.36 dolphins per km. These rates survival in the Karnaphuli-Sangu system are probably are fairly high when compared with other areas of accidental entanglement in monofilament gillnets, bio- shushuk distribution. Dolphin movements in the Sikal- accumulation of persistent contaminants and possibly baha-Chandkhali Canal were consistent with it being collisions with motorized vessels and a decline in prey used as a corridor for migration and dispersal between as a result of over fishing. The most significant conser- the Karnaphuli and Sangu. Shushuks were also sighted vation measure that could be taken would be to in marine waters of the Karnaphuli and Sangu river establish a protected area for dolphins in the Sangu mouths, adding credibility to the hypothesis that River below the Dohazari Bridge. dolphins move along the coast between the Ganges- Brahmaputra-Meghna and Karnaphuli-Sangu systems, Keywords Bangladesh, Kaptai, Karnaphuli, Orcaella, perhaps during the monsoon when freshwater plumes Platanista, river dolphins.

been infrequently reported from Kaptai Lake, one of the Introduction largest reservoirs in Asia, created in 1961 by the Platanista gangetica or shushuk, as the dolphin is called construction of a high dam about 80 km upstream of in Bangladesh, is an endangered river dolphin (Interna- the Karnaphuli river mouth. tional Union of Conservation of Nature (IUCN), 1996) The first record of shushuks in the Karnaphuli River distributed primarily in the Ganges-Brahmaputra-Meg- (and Halda tributary) was a range map contained in hna river system of Nepal, India and Bangladesh (Fig. 1; Anderson (1879). The next reference was a report that Mohan et al., 1997; Smith et al., 1998; Sinha et al., 2000) shushuks were frequently observed close to the mouth but also in the comparatively much smaller Karnaphuli of the Karnaphuli at low and high tides (Pelletier & and Sangu rivers of southern Bangladesh (Fig. 2; Reeves Pelletier, 1986). A photograph of a shushuk specimen, et al., 1993; Ahmed, 2000). Although different water- which was reportedly caught in the vicinity of Chitta- sheds feed the Karnaphuli and Sangu, the Sikalbaha- gong in June 1978, probably from the Halda River Chandkhali Canal connects them in their lower reaches. (Aminul Haque, pers. comm.), was shown in Reeves & Both the rivers and the canal are, hereafter, referred to as Brownell (1989). Reliable sources reported sighting two the Karnaphuli-Sangu complex. Shushuks have also dolphins in the western arm of Kaptai Lake in July 1992 (Reeves et al., 1993). A dead shushuk was found floating Brian D. Smith (corresponding author) Aquatic Biodiversity in the western arm of Kaptai Lake near Rangamati with Associates, PO Box 3479 Eureka California 95502, USA. a fishing rope twisted around its body in October 1994 E-mail: [email protected] (Ahmed, 2000). The existence of a photograph of a Benazir Ahmed Department of Zoology, University of Chittagong, shushuk reportedly found stranded and killed by Chittagong, Bangladesh villagers along the Sangu River, was documented in

Muhammad Edrise Ali Department of Chemistry, Chittagong College, Reeves et al. (1993). Chittagong, Bangladesh The need for a systematic survey of shushuks in Gill Braulik Ocean Park Conservation Foundation, Ocean Park, Kaptai Lake and the southern rivers of Bangladesh was Aberdeen, Hong Kong recognized by the Asian River Dolphin Committee Revised manuscript accepted for publication 9 August 2000 during meetings in 1994 (Reeves & Leatherwood, 1995)

Downloaded from https://www.cambridge.org/core. IP address: 170.106.202.58, on 01 Oct 2021 at 12:09:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1046/j.1365-3008.2001.00153.x 62 B. D. Smith et al.

82°E| 90 °E

30°N r "} 30° N\ Nt TIBET 1 Q) r -"" 1

NEPAL ""\,-'^.._-.. NV BHUTAN w ; r' INDI/k / Southern rivers in survey s ^ i K INDIA 1 V 22°N 22° N c MYANMAF\ 82° E 90"E | \

Fig. 1 The Ganges-Brahmaputra-Meghna and Karnaphuli-Sangu river systems in Nepal, India and Bangladesh.

and 1997 (Smith & Reeves, 2000a). Casual surveys observers searched for dolphins forward of the bow conducted in April and May 1998 (see Ahmed, 2000) whereas another observer, if available, searched revealed a more widespread and abundant population towards the rear. The eye height of observers above than perhaps expected. The surveys also highlighted the waterline was 2-3 m, except during the Bagkhali questions, whether dolphins move between the Sangu and Matamuhuri surveys and Halda survey between and Karnaphuli rivers via the Sikalbaha-Chandkhali Nazirhat Bazaar and Sattar Ghat, when it was approxi- Canal, and among the lower Meghna, Karnaphuli, mately 1 m. In wide channels - defined as being wide Sangu and possibly Matamuhuri and Bagkhali rivers enough so that dolphins surfacing within the observers' (Fig. 3) via coastal waters, maybe during the flood field of view could easily be missed - we followed a season when the rivers' freshwater plumes extend into search path 50-100 m offshore of the riverbank and the Bay of Bengal. arbitrarily alternated between sides. In narrow channels, we followed a search path in the approximate centre of the river channel, except when shallow water Methods required us to follow the thalweg (path connecting the We conducted visual surveys for shushuks in Kaptai deepest points in a series of cross-sections). We main- Lake, the Sikalbaha-Chandkhali Canal, and the Kar- tained a record of search effort, including information naphuli (and Halda tributary), Sangu, Matamuhuri and on sighting conditions, channel type and width, shore- Bagkhali rivers (southern rivers dolphin survey area on line features, human activities, distance covered and the Figs 1-3, respectively) using local motorized and oar- amount of time spent searching for dolphins. During powered boats during January to April 1999. The dolphin sightings, we recorded best, high and low survey was intentionally conducted during the low estimates of group size and estimated the distance water season when the river is accessible and dolphins dolphins were located from the shore. We defined a are more concentrated and therefore more likely to be dolphin group as the group of animals located in the seen. In January 1999, we also conducted a survey along zone of hydrological influence (i.e. river reach) where the coast from Cox's Bazaar to Chittagong searching for the sighting was made. Although we used standardized shushuks and marine cetaceans. criteria for defining river reaches (see Frissel et al., 1986; Kellerhalls & Church, 1989) and attempted to be as We generally followed the survey procedures des- consistent as possible, the definition of reach borders cribed in Smith & Reeves (2000b). Two or three primary

) 2001 FFI, Oryx, 35(1), 61-72

95°45'| 92°00' 92-15'

•• ••• i fOv>"Pablakhali 0 20 40 km 23W I 23°00' / Kaptai f* s^WI Myanimukh Lake \

Mahalchhariaj -Chhota Nt

\nNazirhat vBazaar Rangamati£ ] Kalurghata Halda River ^ Baiukhaii!

Kaptai Hydra-' , I a Sattarghat electric Station^ 23°30' ) Kaptai Dam 23°30' ^Chandraghona Kalurghat Karnaphuli River \Chittagong/ Port Sikalbaha- City or village Chandkhali Canal Bridge " Bay of Bengal

) Sangu Fig. 2 Kaptai Lake, Karnaphuli and Sangu River Dolu River rivers and the Sikalbaha-Chandkhali 95-45' 92°15' Canal. 92°00'

invariably required some subjective judgement in the field. We considered it important, however, for the Results focus of the sighting to be on the animal group vs. No dolphins were observed during our survey of Kaptai individuals. This allowed us to evaluate abundance in Lake. The coverage was fairly widespread and included terms of a range of estimates, rather than an absolute transect lines, all starting from Rangamati, in the count, which would not reflect the inherent uncertainty western arm to Mahalchhari, in the eastern arm to about the actual number of animals present (Smith & Pablakhali and in the far eastern minor arm to Chhota Reeves, 2000b). We avoided double counts by main- Harina. Dolphins were recorded in fairly high densities taining close communication among observers and, for during surveys of the entire Karnaphuli-Sangu complex some sightings, using a zero for our low, or even best, (125 dolphins in total; 0.76 dolphins per km), especially group size estimates, if there was a possibility that the in the Sangu below the Dohazari Bridge (59 dolphins in animals had already been counted. We calculated an total; 1.36 dolphins per km; Table 1). No dolphins were encounter rate for each waterway as the sum of best observed in the Matamuhuri and Bagkhali rivers, but we estimates for all sightings made by the primary received conflicting reports from local fishermen that observers divided by the total distance surveyed. This they occurred far upstream in the Matamuhuri River sum, plus the number of animals sighted by the rear during the monsoon season and in the mouth of the observer that were missed by the primary observers, Bagkhali during the low-water season (Table 2). was used to estimate the minimum abundance of On 31 January, we surveyed 94.0 km along the dolphins in each waterway. coast between Cox's Bazaar and Chittagong, generally

92°10'E

21°50'N 21°50'N

21°40'N 21°40'N 0 5 10 km

City or village B

Bridge Closure Dam C 21 °30'N 21°30'N

Maheshkhali \ j ^ ^P Jhilonja Channel \ Cox's V\> Ramuf .Bazaar" Bay of Bengal Fig. 3 The Bagkhali and Matamuhuri riv- 91°50'E 92°10'E ers showing locations of closure dams.

remaining 1-3 km offshore. We had two sightings of low because of the vastly increased surface area of the Irrawaddy dolphins Orcaella brevirostris for a total of 4-6 reservoir relative to the river before dam construction. dolphins. Both sightings were located less than 2 km We believe, however, that if the lake contained a viable offshore of mangrove forests and in the vicinity of population, our extensive survey effort should have bottom-set gillnets. We also recovered the floating yielded at least a few sightings. carcass of a neonate Irrawaddy dolphin. Visible foetal If shushuks were able to move freely between the folds and a dorsal fin that was not yet fully erect Karnaphuli and Sangu rivers via the Sikalbaha-Chandk- indicated its young age. The total length of the dolphin, hali Canal, the increase in aggregate habitat would measured from the anterior end of the mouth to the provide a greater range of environmental conditions for median notch of the flukes was 105 cm. the dolphins to utilize during different hydrological and tidal conditions. It would also provide greater options for refuge during local anthropogenical disturbances. Discussion Sufficient water was present during a significant portion of the tidal cycle to permit dolphin movement through Distribution and abundance the canal. Observations of shushuks clumped in relat- We are concerned about the absence of dolphin sight- ively high densities at both ends of the canal during low ings in Kaptai Lake. Sighting rates were expected to be tide but distributed more continuously during high tide

)2001 FFI, Oryx, 35(1), 61-72

suggest that the canal may be used as a corridor for best estimates of primary observer sightings for the migration and dispersal. We therefore tentatively con- January surveys of the Karnaphuli and Sangu rivers sider dolphins in the Karnaphuli and Sangu to consti- and the first survey of the Sikalbaha-Chandkhali Canal tute a single population. A caveat is that no dolphins divided by the total survey range where dolphins were were observed in 7.4 km of the middle portion of the observed) and 1.36 dolphins per km in the lower Sangu canal during all three surveys. Although this lends a are comparable or substantially higher than those degree of uncertainty to our reasoning and points to the recorded in other areas of shushuk distribution. A importance of further investigation, one of the authors downstream survey of the Vikramshila Gangetic Dol- (B.D.S.) has observed shushuks occupying a shallower phin Sanctuary in the middle reaches of the Ganges and narrower channel in the Kulsi River, a small mainstem (60.3 km covered) recorded 0.81 dolphins tributary of the Brahmaputra in India (see also Mohan per km (Sinha et al., 2000). Encounter rates have been et al., 1998). reported as generally lower in other river segments, The minimum abundance-estimate for the Karnaph- such as 0.24 dolphins per km in the middle reaches of uli-Sangu complex (125 dolphins) was based on the sum the Jamuna (Brahmaputra) River (89.6 km surveyed) of best estimates for the January surveys of both rivers and 0.30 dolphins per km in the Kalni-Kushiyara River (92 dolphins seen by primary observers plus seven seen (113.0 km surveyed) - rates recalculated from sigh- by the rear observer), the sum of estimates of the first tings per km reported in Smith et al. (1998). The Sangu survey of the Sikalbaha-Chandkhali Canal (18 dol- River downstream of the Dohazari Bridge supported phins), and the number of animals observed during the highest density of dolphins and, to the best of the the survey of the Halda River in February (four dolphins authors' knowledge, the encounter rate recorded in seen by the primary observer and four seen by the rear this segment is the highest documented for the observer). Due to biases associated with estimating river species. The lower Sangu is the most undisturbed dolphin abundance from direct counts (e.g. long sub- area within the range of dolphins in the Karnaphuli- mergence times, cryptic surfacings and observer inat- Sangu complex and may serve as an important refuge tention or focus on locations different from where from environmental disturbances, such as industrial dolphins are surfacing (see Smith & Reeves, 2000b), pollution and high vessel traffic, in the lower Karna- we believe that the actual number of dolphins in the phuli. Karnaphuli-Sangu complex could be substantially The absence of dolphin sightings in the Matamuhuri higher. and Bagkhali rivers may have been a result of the The rear observer observed 10-32 per cent of the total construction of seasonal closure dams, which block all number of dolphins sighted. This does not include the freshwater outlets during the dry season. Reports of Halda survey, where we had only a single primary dolphins occurring upstream of the dam sites in the observer and used a small rowboat for a large portion of Matamuhuri River during the monsoon and the recov- the survey. In this narrow channel, the rear observer ery of a lower jaw of a shushuk, reportedly caught saw 50 per cent of the total number of dolphins sighted. below the Bagkhali Rubber Dam in January, indicate Dolphins were also occasionally seen after survey effort that shushuks may have historically used these rivers was suspended but then not observed by the rear or during the dry season, and may still use them during the primary observers when we resumed search effort a few flood season. Assuming that shushuks have the ability minutes later. This indicates that direct counts under- and motivation to disperse along the coast of southern estimate absolute abundance, which was expected con- Bangladesh, the blockage of the Matamuhuri and sidering sighting biases. We believe, however, that our Bagkhali rivers during the dry season would then estimates of minimum and relative abundance provide represent a reduction in freshwater habitat. important information for assessing the viability of the Sightings of shushuks in marine waters of the Karnaphuli-Sangu population and serve as baseline Karnaphuli and Sangu river mouths indicate that they data for detecting long-term abundance trends (see can tolerate saline environments for at least a short Taylor & Gerrodette, 1993), however, for a discussion of period. The limitation of their range to freshwater problems with detecting population trends in small bodies, therefore, may be better explained by ecological populations). Additional data from rear observer sight- adaptations and inter-specific competition, possibly ings may allow for a correction factor to be calculated to with Irrawaddy dolphins, rather than physiological reduce the negative bias of minimum estimates obtained requirements. These observations also imply that when a rear observer is not available (see Smith & shushuks may be able to move along the coast of the Reeves, 2000b). Bay of Bengal among the lower Meghna River, Karna- The encounter rates of 0.76 dolphins per km in the phuli-Sangu complex, and Matamuhuri and Bagkhali Karnaphuli-Sangu complex (calculated from the sum of rivers.

) 2001 FFI, Oryx, 35(1), 61-72

175 3

. IPaddress: Table 1 Survey details for Kaptai Lake and Karnaphuli, Halda and Sangu rivers.

Distance (km) Sightings Sightings Group size Location Date (Vessel speed (km (Primary observers) (Rear observer) (dolphins) 170.106.202.58

Kaptai Lake 23-27 Feb 224.8 (8.7) None None Karnaphuli River from the 22-23 Jan 70.8 (8.7) 19 sightings of 33-37-28° No rear observer used Mean = 1.4 Kaptai Hydroelectric Station dolphins; 0.47 dolphins per km*1 (SD = 0.8, range = 1-1) , on (c. 5 km below Kaptai Dam) 01 Oct2021 at12:09:21 to the river mouth

Karnaphuli River from the 23-24 Apr 72.5" (11.5) 14 sightings of 11 sightings of Mean = 1.9 Kaptai Hydroelectric Station 27-32-24 dolphins; 13 dolphins (32%)d (SD = 1.3, range = 1^1) to the river mouth 0.32 dolphins per km Halda River downstream from 13 Feb 39.5 (5.9) Four sightings of four dolphins; Four sightings of four Mean = 1.0

, subjectto theCambridgeCore termsofuse,available at Nazirhat Bazaar to the 0.10 dolphins per km dolphins (50%) (SD = 1.0) confluence of Karnaphuli

Sangu River from 30 km above the 25-26 Jan 83.5 23 sightings of 59-71-52 dolphins; Seven sightings of seven Mean = 2.6 Dohazari Bridge at the start of the (6.8 above the 0.71 dolphins per km dolphins (11%) (SD = 2.6, range = 1-12) Bandarban hills to the river mouth bridge and 9.9 but 1.36 dolphins per km within below the bridge) 43.5 km of their range Sangu River from Dohazari 30 Apr 48 (9.2) 17 sightings of 33-44-29 dolphins; Seven sightings of Mean = 1.9 Bridge to the river mouth 0.69 dolphins per km 14 dolphins (30%) (SD = 1.4, range = 1-6) Sikalbaha-Chandkhali Canal 24 Jan 29.4 (7.2) 14 sightings of 18-20-17 dolphins; Two sightings of two Mean = 1.3 from Karnaphuli to Sangu 0.61 dolphins per km dolphins (10%) (SD = 0.5, range = 1-2) Sikalbaha-Chandkhali Canal 27 Jan 29.4 (9.4) 15 sightings of 27-28-26 dolphins; Three sightings of three Mean = 1.8 from Sangu to Karnaphuli 0.92 dolphins per km dolphins (10%) (SD = 1.1, range = 1^) Sikalbaha-Chandkhali Canal 27 Apr 29.8 (12.4) 14 sightings of 20-24-18 dolphins; Four sightings of four Mean = 1.4 from Karnaphuli to Sangu 0.67 dolphins per km dolphins (17%) (SD = 0.64, range = 1-3) https://www.cambridge.org/core/terms . https://doi.org/10.1046/j.1365-3008.2001.00153.x Downloaded from https://www.cambridge.org/core

Table 1 (Continued).

Location Distance from shore (m) Comments

Kaptai Lake Fishermen at Rangamati reported occasionally observing shushuks near Barkal during August-September and near Kalurghata and Balukhali during March-April. . IPaddress: Karnaphuli River from the Mean = 138 Farthest upstream sighting was c. 10 km below Kaptai Dam. Farthest downstream sighting was located Kaptai Hydroelectric Station (SD = 206, range = 20-1000) at the river mouth in saline water during a mid-rising tide (low 0.2 m; high 3.9 m). (c. 5 km below Kaptai dam) 170.106.202.58 to the river mouth Karnaphuli River from the Mean = 95 Farthest upstream sighting was c. 8 km below Kaptai Dam. Farthest downstream sighting was c. 1.5 km Kaptai Hydroelectric Station (SD = 57, range = 10-200) from the river mouth in saline water during a late mid-falling tide (low 1.6 m; high 3.5 m). to the river mouth , on

01 Oct2021 at12:09:21 Halda River downstream from Mean = 9 An oar-powered dinghy was used from Nazirhat Bazaar Ghat to Sattar Ghat as a result of shallowness Nazirhat Bazaar to the (SD = 4, range = 3-15) of the water. Only one primary observer and one rear observer were used on both vessels. confluence of the Karnaphuli Sangu River from 30 km above the Mean = 80 Progress slow upstream of the bridge because of low water. Farthest upstream sighting was below the bridge Dohazari Bridge at the start of the (SD = 96, range = 10^00) and c. 44.6 km upstream of the mouth. Farthest downstream sighting was in the river mouth in saline water Bandarban hills to the river mouth during an early mid-rising tide (low 0.5 m, high 3.6 m).

, subjectto theCambridgeCore termsofuse,available at Sangu River from Dohazari Bridge Mean = 66 Survey ended c. 2 km upstream from the mouth because of water turbulence. Farthest upstream sighting to the river mouth (SD = 37, range = 5-100) was c. 43.7 km upstream of the mouth. Farthest downstream sighting was at the survey endpoint during an early mid-falling tide (low 0.9 m, high 4.4 m). Sikalbaha—Chandkhali Canal Mean = 21 Vessel became grounded 18.1 km into the canal, less than 1 h before low tide (0.3 m) We were unable to resume from Karnaphuli to Sangu (SD = 5.7, range = 10-30) until almost 7 h later when the canal filled from the Karnaphuli side c. 2 h before high tide (3.7 m). Sikalbaha-Chandkhali Canal Mean = 19 Vessel had clear passage through the canal during an early mid-falling tide, passing the point where we became from Sangu to Karnaphuli (SD = 6.0, range = 10-30) grounded on the first survey during late mid-falling tide (low 0.7 m; high 3.6 m). Q. O_ Sikalbaha-Chandkhali Canal Mean = 17 Vessel had clear passage through the canal during a mid-rising to high tide (low 0.9m, high 4.1m). T3 from Karnaphuli to Sangu (SD = 7.0, range = 8-30) o "Sum of best-high-low estimates. ^Encounter rates calculated from sum of best estimates. O Difference in distances covered during January and April surveys was because of the slightly different routes travelled within the same channel. ^Percentage of the total number of observed dolphins (sum of best estimates of primary and rear observers) seen by the rear observer.

https://www.cambridge.org/core/terms 0) (Q_ 0) a. CD w . https://doi.org/10.1046/j.1365-3008.2001.00153.x Downloaded from https://www.cambridge.org/core

•D

CO 3

Table 2 Survey details for the Matamuhuri and Bagkhali rivers. . IPaddress: Distance (km) Location Date (Vessel speed (km h"1)) Comments

170.106.202.58 Matamuhuri River from 4.2 km upstream 30 Jan 33.7 (7.7) Conflicting reports of shushuks occurring as far upstream as Manikpur during the of Manikpur downstream to the Furityakhali monsoon season. Furityakhali closure dam is c. 50 m long and 3 m high, and made of closure dam in the northern channel of the mouth sticks and mud. Its purpose is to prevent saltwater intrusion so that river water can be 8.9 km downstream of the bifurcation used for irrigation. The dam is rebuilt each year after being destroyed by annual floods. , on Matamuhuri River, southern channel from the mouth 18 Mar 17.1 (7.4) Local fisherman reported that he often sees dolphins downstream of the dam during high 01 Oct2021 at12:09:21 to the Rampur earth closure dam tide but his description resembled Irrawaddy dolphins Orcaella brevirostris. The purpose and construction of the Rampur closure dam are the same as the Furityakhali closure dam. Matamuhuri River, northern channel from the mouth 19 Mar 43.4 (12.2) The purpose and construction of the Pekua (spelled Bekua in Smith et ah, 2000) closure and through the Bhola Canal to the Fekua closure dam are the same as the Furityakhali and Rampur closure dams. dam (13.2 km), back down the Bhola Canal and up

, subjectto theCambridgeCore termsofuse,available at the channel to the Furityakhali closure dam (14.3 km) and downstream to the river mouth (15.9 km)

Bagkhali River upstream from the river mouth until 29 Jan 10.4 (10.1) We received a report that two shushuks were killed accidentally in a gillnet just below the the Bagkhali Rubber Dam at Jhilonja dam about 20 days prior to our visit. The fisherman, who caught the dolphins, described accurately the long rostrum, sharp protruding teeth and very small dorsal fin characteristic of shushuk. He said that the dolphins were present just downstream of the dam for several days before being caught. Apparently, a local villager took the carcasses for extracting oil. We were unable to find this person but another villager gave us the lower jaw of a shushuk that he claimed was caught in January 1998 at the same location. The teeth on the lower jaw were worn at the tips, especially those at the distal end, indicating that it was from an older animal.

Bagkhali River from Ramu Bazaar downstream to the 20 Mar 17.6 (3.6) An oar-powered vessel was used for the survey as a result of shallow water. Dam is a Bagkhali Rubber Dam at Jhilonja water-filled rubber tube 100-120 m long and about 4 m tall. Bagkhali River from the Bagkhali Rubber Dam to the 21 Mar 10.9 (9.6) None river mouth https://www.cambridge.org/core/terms . River dolphins of southern Bangladesh 69

River and canal confluences. Mosquito nets were com- Habitat preferences monly observed set in shallow channels to catch fish fry Shushuk habitat is generally described as consisting of and crustacean larvae for aquaculture purposes. These eddy counter-currents located downstream of channel nets are particularly destructive when set during convergences and sharp meanders and upstream and breeding migrations and early growth of fishes and downstream of mid-channel islands (Smith, 1993; Smith crustaceans. et al., 1998; Smith & Reeves, 2000c). It has been sugges- Concentrations of organochlorines have declined in ted that the affinity of shushuks for counter-currents some cetacean populations as certain countries restric- declines in large channels where current velocity is ted their production and use (see O'Shea et al., 1999), but reduced and eddy boundaries are ill-defined (Smith in other countries, particularly developing ones, many et al, 1998). Shushuks in the Sangu and Karnaphuli harmful contaminants continue to be manufactured and appeared to be attracted to counter-currents when used without regard to their potential effects on people current velocity was high, which depended on the tidal and wildlife. Untreated contaminants are openly dis- cycle, but became more uniformly dispersed during charged from the Karnaphuli Paper and Rayon complex slack water. In April 1998, during a casual survey, two at Chandroghona. Pesticides, including some that of the authors (B.D.S. and B.A.) observed a very large have been banned in more developed countries are group of about 20 shushuks in a counter-current widely used in shoreline agriculture. Organochlorines downstream of a sharp meander in the lower reaches contained in pesticides and industrial chemicals can of Karnaphuli during a rapidly falling tide. Dolphins bioaccumulate over the lifetime of individuals and from were absent from this area during our first Karnaphuli one generation to the next, via nursing and transfer survey at mid-rising tide and present in groups of five across the placenta, to levels disastrous for the long-term during the second survey at mid-falling tide. Dolphins survival of dolphins (O'Shea et al, 1999). A study of were also observed in the tidal lee of anchored vessels at shushuks in the Ganges River found that the residue Chittagong Port. Tidal and river flow conditions, inclu- levels of organochlorines were high and that the ding the hydrological modifications caused by Kaptai dolphins had a lower capacity to metabolize organo- dam and the trans-flow between the Karnaphuli and chlorines compared with small marine cetaceans Sangu via the Sikalbaha-Chandkhali Canal, create a (Kannan et al, 1993, 1994). River dolphins probably also complex aquatic environment. Dolphin distribution face a greater risk from environmental contamination within the Karnaphuli-Sangu complex probably reflects because their habitat in counter-currents is often located the patchy and dynamic distribution of prey and at pollution discharge sites (Kannan et al, 1994) and the hydraulic refuge. hydraulics of these areas may inhibit the dispersal of pollutants. Collisions with large and small vessels in the conges- Threats ted Chittagong Port may be a source of shushuk Similar to all cetaceans, shushuks are vulnerable to mortality, especially during calving and nursing periods accidental entanglement in monofilament gillnets. when the dolphins' ability to evade boats is compro- According to local fishermen in the Brahmaputra River mised. The size of the port and the number of mechan- near Sirajganj, Bangladesh, accidental entanglements in ized vessels are likely to increase as the economy of monofilament gillnets are their primary source of Bangladesh develops. A shushuk was reported struck dolphin oil, which is used as a fish attractant (see and killed by a large cargo vessel in the Brahmaputra Motwani & Srivastava, 1961) and accounts for mortality River near India-Bangladesh border (Mohan, 1996). in the middle reaches of the river that was roughly estimated to be 90-160 dolphins each year (Smith et al, 1998). Monofilament gillnets (current-jahls) with a Conservation stretched mesh-size of less than 4.5 cm are prohibited A potentially viable population of shushuks is now in Bangladesh, but their use is increasing (Liquat Ali, known to occur in the Karnaphuli-Sangu complex. personal communication). Shushuks have been reported Although we have no proof of demographic interaction to be caught in jam jahls, rectangular nets with an between shushuks in the Karnaphuli-Sangu and 8-10-cm mesh size used to catch large economically Ganges-Brahmaputra-Meghna systems, our findings valuable fish in counter-currents (Bangladesh Water provide evidence that supports the possibility. Dispersal Development Board, 1994). Both these types of nets were between these systems does not imply, however, that observed being used in the Karnaphuli, especially near the species can be considered a single evolutionary the canal confluence and in the large counter-current unit (see Dizon et al, 1992). Even with substantial close to the mouth, and in the Sangu, near the Dolu interchange of individuals, genetic differences between

populations can be considerably different because of 3 Water quality be investigated and monitored on a selection and genetic drift (Slatkin, 1987). regular basis. Special emphasis should be paid to We reiterate the suggestion in Smith et al. (1998) that persistent contaminants, particularly organochlorines freshwater dolphins in Bangladesh deserve conservation such as polychlorobiphenyl (PCB) and dichlorodiphe- attention not because they are in immediate danger of nyltrichloroethane (DDT). Dolphin blubber and liver extinction but because they still occur in sufficient samples obtained from strandings or accidental kills can numbers for early conservation efforts to be effective. be used to assess the types, levels and health effects of In contrast to encouraging results of dolphin surveys in contaminants passed through the riverine food chain. the Karnaphuli-Sangu complex, the negative results of With clean stainless steel instruments, a large sample of our surveys in Kaptai Lake should be considered a blubber (e.g. 10 x 10 cm) should be collected from the warning to resource managers about the ability of river mid-thoracic area and a small sample of liver collected dolphins to adapt to the lacustrine conditions of a from the left lobe (Geraci & Lounsbury, 1993; Clausen, reservoir environment. 1999). Ideally, the samples would be immediately placed on ice in a Teflon-coated plastic bag for transport and then stored in a freezer at -70 °C (Geraci & Lounsbury, Conservation recommendations 1993). If a freezer is not available, the samples should be Balancing the needs of wildlife and human beings is a stored in a sterilized glass jar with a piece of aluminium difficult task and this applies particularly to river foil placed between the lid and sample, and kept in a dolphins that share freshwater resources with one of cool place (P.J.H. Reijnders, pers. comm.). the world's most densely populated and economically 4 The abundance and habitat of shushuks be monitored impoverished human populations. Nevertheless, we on a regular basis using standardized techniques. Long- believe that the environmental requirements of river term monitoring is required to detect population trends dolphins reflect the survival needs of local people - a and to assess habitat changes caused by human activit- sufficient supply of reasonably unpolluted freshwater ies. Comparisons of relative abundance and habitat use and a sustainable source of high quality food - and that should be made only among standardized surveys this linkage should be used as a basis for implementing conducted during similar river flow and tidal stages, effective conservation. preferably when water levels are at their lowest (see We suggest that the following initiatives, listed in Smith & Reeves, 2000b). Surveys conducted during rough order of priority, be taken to protect the long-term different flow and tidal conditions would be useful, viability of the Karnaphuli-Sangu dolphin population. however, for understanding how shushuks use habitat 1 A protected area for shushuks be established in the during different flow and salinity conditions. Detailed Sangu River below the Dohazari Bridge. Establishing monitoring of a few selected habitat sites could involve meaningful protection for dolphins will require imple- systematic studies of food availability (catch-per-unit menting a conservation plan that has the support of effort for fish and crustaceans), human activities (fishing, local people and government officials. A 'managed transport, industry, agriculture, etc.), and physical (mor- resource protected area' (IUCN, 1994) where local phology, salinity and current speed and distribution) communities manage fisheries for sustainable use might and biological (productivity, and indices of fish and be supported by local people and therefore viable in the benthic invertebrate diversity) conditions. A population long term, if adequate protection for river dolphins and and habitat monitoring programme should involve their prey can be assured (see Smith & Reeves, 2000a). students from Chittagong University and rangers from The most important step would be to reduce sources of the Bangladesh Forest Department and entail a strong accidental killing in gillnets and protect fish populations training component conducted prior to survey activities. by restricting or eliminating the use of non-selective 5 An investigation be conducted on demographic fishing gear (monofilament gill- and mosquito nets) and interaction between shushuk populations in the promoting more traditional techniques (hook and line, Ganges-Brahmaputra-Meghna and Karnaphuli-Sangu and throw and lift nets). river systems. The study should combine information on 2 An awareness programme be initiated to educate shushuk genetics and morphology. For genetic studies, a local people and government officials on the import- small piece of skin should be collected from recovered ance of conserving river dolphins and maintaining carcasses and stored in a glass or plastic vial filled with sustainable fisheries. The programme should use 20 per cent dimethyl sulphoxide (DMSO) and saturated popular media, such as video, posters, comic books, sodium chloride. Storage vials should be clearly labelled radio and street theatre; be interactive and respectful with the collection date and location, sex if known, and of local values and traditions; and involve local name of the collector. Cross-contamination must be university students as environmental educators. avoided by using sterilized instruments only (i.e.

>2001 FFI, Oryx, 35(1), 61-72

washed in soapy water and rinsed in ethanol). In Reijnders, A. Aguilar and G. P. Donovan), pp. 43^16. IWC, decreasing order of usefulness dried skin, teeth and Cambridge, UK. Dizon, A.E., Lockyer, C, Perrin, W.F., DeMaster, D.P. & Sisson, bones can also be used. A small sample of these should J. (1992) Rethinking the stock concept: a phylogeographic be collected and kept dry individually, in clean plastic approach. Conservation Biology, 6, 24-36. bags (Rosel & Reeves, 2000). For morphological studies, Frissel, C.A., Liss. W.J., Warren, C.E. & Hurley, M.D. (1986) standard measurements of available carcasses and A hierarchical framework for stream habitat classification: skeletal materials should be taken, ideally by a single viewing streams in a watershed context. Environmental researcher in order to reduce variability. Information Management, 10, 199-214. from bioacoustics, organochlorine levels, parasite spe- Geraci, J.R. & Lounsbury, V.J. (1993). Marine Mammals Ashore: a Field Guide for Strandings. Texas A & M Sea Grant Publication, cies, and perhaps telemetry studies can also be used Galveston, Texas. (Smith & Reeves, 2000a; Appendix 6). Expertise for IUCN (1994) Guidelines for Protected Area Management Categories. conducting these studies is available through the IUCN IUCN Commission on National Parks with assistance of the Species Survival Commission's Cetacean Specialist World Conservation Monitoring Centre, IUCN, Gland, Group. Although some analyses may have to be Switzerland. conducted outside Bangladesh, close collaboration IUCN (1996) 1996 IUCN Red List of Threatened Animals. IUCN, should be maintained with local scientists. Gland, Switzerland. Kannan, K., Sinha, R.K., Tanabe, S., Ichihashi, H. & Tatsukawa, R. (1993) Heavy metals and organochlorine residues in Acknowledgements Ganges river dolphins from India. Marine Pollution Bulletin, 26, 159-162. This investigation would not have been possible without Kannan, K, Tanabe, S., Tatsukawa, R. & Sinha, R.K. (1994) the financial support from the Ocean Park Conservation Biodegradation capacity and residue pattern of organochlo- Foundation and the Whale and Dolphin Conservation rine in the Ganges river dolphins from India. Toxicological and Society. Special thanks are given to Thomas Jefferson for Environmental Chemistry, 42, 249-261. championing the project and to Carmen Lor and Kellerhalls, R. & Church, M. (1989) The morphology of large Josephine Woo and Alison Smith for organizing the rivers: characterization and management. In Proceedings of the International Large River Symposium, Vol 106 (ed. D. P. Dodge), funds. We would also like to thank the numerous pp. 31^18. Canadian Special Publications and Aquatic boatmen who assisted us during the surveys and Dr Sciences, Ottawa, Canada. Irshad Kamal Khan and Dr Shafique Haider Chowdh- Mohan, R.S.L. (1996) Rivdolphinews (Newsletter of the Ganges ury for lending us their e-mail facilities. Appreciation is River Dolphin Protection Committee), 1, 1-10. given to Manmatha Nath Sarker for helping us arrange Mohan, R.S., Dey, S.C., Bairagi, S.P. & Roy, S. (1997) On a boats and lodging when we were in Cox's Bazaar. Dr survey of the Ganges river dolphin Platanista gangetica of Farid Ahsan, Ripon Chakma, Abdul Hannan and Kam- Brahmaputra River, Assam, journal of the Bombay Natural rul Hasan are gratefully acknowledged for their assist- History Society, 94, 483^95. Mohan, R.S., Dey, S.C. & Bairagi, S.P. (1998) On a resident ance with fieldwork. We thank Dr William Perrin and an population of the Ganges river dolphin Platanista gangetica in anonymous reviewer for their useful comments on the the Kulsi River (Assam), a tributary of Brahmaputra. Journal manuscript. of the Bombay Natural History Society, 95, 1-7. Motwani, M.P. & Srivastava, C.B. (1961) A special method of fishing for Clupisoma garua (Hamilton) in the Ganges river References system, journal of the Bombay Natural History Society, 58, 285-286. Ahmed, B. (2000) Water development and the status of the O'Shea, T.J., Reeves, R.R. & Long, A.K. (eds) (1999) Marine shushuk (Platanista Gangetica) in Southeast Bangladesh. In Mammals and Persistent Ocean Contaminants. Proceedings of Biology and Conservation of Freshwater Cetaceans in Asia, the Marine Mammal Commission Workshop Keystone, Colorado, Occasional Papers of the IUCN Species Survival Commission 12-15 October 1998. Marine Mammal Commission, Bethesda, no. 23 (eds R. R. Reeves, B. D. Smith and T. Kasuya), Maryland. pp. 62-66. IUCN, Gland, Switzerland. Pelletier, C. & Pelletier, F.X. (1986) Le Plataniste du Gange. Anderson, J. (1879) Anatomical and Zoological Researches: Com- Bhulan, dauphin sacre. L'univers Du Vivant, 8, 8-18. prising an Account of the Zoological Results of the Two Expedi- Reeves, R.R. & Brownell, R.L. Jr (1989) Susu Platanista tions to Western Yunnan in 1868 and 1875; and a Monograph of gangetica (Roxburgh, 1801) and Platanista minor Owen, the Two Cetacean Genera Platanista and Orcaella. Quaritch, 1853. In Handbook of Marine Mammals, Vol 4 (eds S. H. London. Ridgeway and R. Harrison), pp. 69-99. Academic Press, Bangladesh Water Development Board (1994) Northeast Regional London. Water Management Project (PAP 6), Fisheries Specialist Study. Reeves, R.R. & Leatherwood, S. (1995) Report of the First Meeting Unpublished report, BWDB, Dhaka. of the Asian River Dolphin Committee, Ocean Park, Hong Kong. Clausen, B. (1999) Practical guidelines for postmortem exam- 5-7 December 1994. Ocean Park Conservation Foundation, ination and tissue sampling of cetaceans for ecotoxicological Hong Kong. purposes. In Chemical Pollutants and Cetaceans (eds P. J. H.

Reeves, R.R., Leatherwood, S. & Mohan, R.S.L. (1993) Report Smith, B.D., Sinha, R.K., Zhou, K. et al. (2000) Register of from the Seminar on the Conservation of River Dolphins of The water development projects affecting Asian river cetaceans. Indian Sub-Continent. 18-19 August 1992, New Delhi, India, In Biology and Conservation of Freshwater Cetaceans in Asia, Whale and Dolphin Conservation Society, Bath, UK. Occasional Papers of the IUCN Species Survival Commission Rosel, P.E. & Reeves, R.R. (2000) Genetic and demographic no. 23 (eds R. R. Reeves, B. D. Smith and T. Kasuya), considerations for the conservation of Asian river cetaceans. pp. 22-39. IUCN, Gland, Switzerland. In Biology and Conservation of Freshwater Cetaceans in Asia, Taylor, B.L. & Gerrodette, T. (1993) The uses of statistical power Occasional Papers of the IUCN Species Survival Commission in conservation biology: the vaquita and northern spotted no. 23 (eds R. R. Reeves, B. D. Smith and T. Kasuya), pp. 144- owl. Conservation Biology, 7, 439-500. 152. IUCN, Gland, Switzerland. Sinha, R.K., Smith, B.D., Sharma, G., Prasad, K., Choudhury, B.C., Sapkota, K., Sharma, R.K. & Behera, S.K. (2000) Status and distribution of the Ganges susu. Platanista Gangetica, in Biographical sketches the Ganges river system of India and Nepal. In Biology and Conservation of Freshwater Cetaceans in Asia, Occasional Papers Brian D. Smith is Asia Coordinator for the IUCN Species of the IUCN Species Survival Commission no. 23 (eds R. R. Survival Commission Cetacean Specialist Group and the Reeves, B. D. Smith and T. Kasuya), pp. 54-61. IUCN, Gland, Whale and Dolphin Conservation Society. He has conducted Switzerland. research on marine and freshwater cetaceans in Asia since Slatkin, M. (1987) Gene flow and geographic structure of 1990 and has published papers on his work in Nepal, India, natural populations. Science, 236, 787-792. Bangladesh, Vietnam and Myanmar. He recently co-edited a Smith, B.D. (1993) 1990 status and conservation of the Ganges special volume on marine mammals for Asian Marine Biology river dolphin (Platanista gangetica) in the Karnali River, and a volume on the biology and conservation of freshwater Nepal. Biological Conservation, 66, 159-170. cetaceans in Asia for the IUCN Species Survival Commission Smith, B.D., Haque, A.K.M., Hossain, M.S. & Khan, A. (1998) Occasional Papers Series. River dolphins in Bangladesh: conservation and the effects of water development. Environmental Management, 22, Benazir Ahmed is Professor of Zoology at the University of 323-335. Chittagong. He has been teaching and conducting research Smith, B.D. & Reeves, R.R. (2000a) Report of the Second on fisheries and wildlife in Bangladesh for over 25 years. Meeting of the Asian River Dolphin Committee, 22-24 River dolphins have been the primary focus of his interest February 1997, Rajendrapur, Bangladesh. In Biology and for the last 10 years and he is a member of the Asian River Conservation of Freshwater Cetaceans in Asia, Occasional Papers Dolphin Committee. of the IUCN Species Survival Commission no. 23 (eds R. R. Muhammad Edrise Ali is a lecturer in chemistry at Chitta- Reeves, B. D. Smith and T. Kasuya), pp. 1-14. IUCN, Gland, gong College. He has conducted research on the environ- Switzerland. mental chemistry of rivers in south-east Bangladesh for the Smith, B.D. & Reeves, R.R. (2000b) Survey methods for last 10 years and been involved in river dolphin research for population assessment of Asian river dolphins. In Biology and the last 3 years. He is a member of the Royal Society of Conservation of Freshwater Cetaceans in Asia, Occasional Papers Chemistry, UK. of the IUCN Species Survival Commission no. 23 (eds R. R. Reeves, B. D. Smith and T. Kasuya), pp. 97-115. IUCN, Gill Braulik has been conducting marine mammal research Gland, Switzerland. in Hong Kong since 1996 and has been the project coordi- Smith, B.D. & Reeves, R.R. (2000c) Report of the workshop on nator for two large marine mammal research projects in the effects of water development on river cetaceans in Asia, Hong Kong, one focused on Indo-Pacific humpbacked Rajendrapur, Bangladesh, 26-28 February 1997. In Biology and dolphins and the other on finless porpoises. She has Conservation of Freshwater Cetaceans in Asia, Occasional Papers conducted surveys, participated in workshops and assisted of the IUCN Species Survival Commission no. 23 (eds R. R. with training courses in Pakistan, USA, Vietnam, China, Reeves, B. D. Smith and T. Kasuya), pp. 15-22. IUCN, Gland, Singapore, Indonesia, Bangladesh and Hong Kong. Switzerland.

>2001 FFI, Oryx, 35(1), 61-72